Each year, more than one million new cases of breast cancer are diagnosed worldwide, and an estimated 370,000 women die from breast cancer. The vast majority of fatal breast cancer cases involve metastatic spread of the primary tumor, but the metastatic process remains a complex and poorly understood process. Our long-range goal is to identify the molecular mechanisms of breast cancer progression from solitary tumor to metastasis. Based on our extensive molecular analysis of clinical human breast cancer specimens and a series of experimental breast cancer models, we now propose to test the following central hypothesis: Loss of activation of transcription factor Stat5 is a cancer progression event that favors epithelial-to-mesenchymal dedifferentiation, invasiveness, and increased metastatic potential of breast cancer cells. To accomplish the objectives of this application, we will pursue three specific aims: Aim #1: Analyze 1,300 human breast cancer specimens to establish the relationship between levels of active Stat5 and measures of tumor cell invasiveness. Aim #2: Establish whether Stat5 stimulates human breast cancer cell differentiation and adhesion, and suppresses tumor cell invasion, in vitro and in vivo. Aim #3: Determine the effect of Stat5 activation on invasion and metastasis of mouse breast cancer models in vivo. Our expectations are that by the end of the proposed project period, we will have established whether Stat5 activation status in breast cancer is a useful clinical predictor of disease progression and clinical outcome in lymph node-negative breast cancer. This is important because active Stat5 may serve as a simple immunohistochemical marker to identify node-negative breast cancer patients with excellent prognosis, permitting more individualized treatment, including selection of antiestrogen therapy. The results of this work could lead to new preventative and therapeutic strategies for primary and metastatic breast cancer.